Perception & Recognition on VLMsAreBiased
77.4Pass@1Seed2.0 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Seed2.0 Pro2026.06 | 77.4 | |
| Seed2.0 Lite2026.06 | 74.8 | |
| Seed1.82026.06 | 62 | |
| Seed2.0 Mini2026.06 | 58.4 | |
| Gemini-3-Pro High2026.06 | 50.6 | |
| GPT-5.2 High2026.06 | 28 | |
| Claude-Opus-4.52026.06 | 21.4 |